Postoperative fibrinogen level is associated with postoperative bleeding following cardiothoracic surgery and the effect of fibrinogen replacement therapy remains uncertain

Authors


  • Manuscript handled by: I. Pabinger
  • Final decision: F. R. Rosendaal, 22 May 2013

Correspondence: Lucy Yang, Addenbrooke's Hospital, Box 110, Level 3 ACCI Building, Hill's Road, Cambridge CB2 0QQ, UK.

Tel.: +44 1223 296163; fax: +44 1223 216893.

E-mail: ly264@medschl.cam.ac.uk

Summary

Background: Traditionally, a fibrinogen level > 1 g L−1 has been viewed as the critical plasma concentration required for hemostasis. No definitive trial has investigated the plasma fibrinogen hemostatic threshold and fibrinogen replacement in complex surgical patients with acquired bleeding. Objectives: To explore the plasma fibrinogen level required for hemostasis in cardiothoracic surgery patients and assess the association of fibrinogen replacement therapy (using cryoprecipitate or fibrinogen concentrate) with reducing postoperative bleeding rate. Patients/Methods: Data from a prospectively collated database were used to examine the relationship between postoperative plasma fibrinogen level and the postoperative rate of bleeding within the hour of plasma fibrinogen measurement (n = 430) and to explore the effect of cryoprecipitate infusion (n = 76) or fibrinogen concentrate administration (n = 8) on postoperative bleeding rate. Results: A low plasma fibrinogen level was significantly associated with bleeding, with an odds ratio of 3.06 for every 1 g L−1 decrease in fibrinogen (95% confidence interval 1.05–8.90) with adjustment for confounders. A fibrinogen threshold associated with excess bleeding was not identified, but this relationship was a continuum. There was no reduction in bleeding following administration of cryoprecipitate or fibrinogen concentrate to raise the post-infusion fibrinogen level to a median of 2.00 and 1.70 g L−1, respectively. Conclusions: There is a continuum of bleeding severity with reducing fibrinogen concentration. Fibrinogen concentrate or cryoprecipitate infusion did not significantly reduce bleeding rate; however, confirmation by a randomized controlled trial is required. It remains uncertain whether low postoperative fibrinogen levels are causally associated with postoperative bleeding.

Ancillary